1 / 22

Reinsurance Institute: Generic Model Development and State Consultation

This presentation provides an overview of the Reinsurance Institute's progress in developing a generic model for reinsurance, as well as its consultation with states. It discusses the modeling approach, policy parameters, and challenges in modeling reinsurance at the state level.

rmaria
Download Presentation

Reinsurance Institute: Generic Model Development and State Consultation

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. The Reinsurance Institute I:Generic Model Development and State Consultation Randall R. Bovbjerg, JD Principal Research Assoc. A. Bowen Garrett, PhD Senior Research Assoc. Lisa Clemans-Cope, PhD Research Associate Health Policy Center, The Urban Institute Presentation to Reinsurance Institute’s Final Meeting with States, Marriott-Metro Center, Washington, DC; Thursday, July 19, 2007 The Reinsurance Institute is funded through the State Coverage Initiatives (SCI) program, which is a national program of the Robert Wood Johnson Foundation, administered by AcademyHealth, Washington, DC. Content of slides is responsibility of the authors.

  2. What is the SCI Reinsurance Institute? SCI/AcademyHealth Enrique Martinez-Vidal, Director SCI Donald Cohn, Alice Burton (now working for MD) ... and participating states Urban Institute HPC & Team Randall R. Bovbjerg A. Bowen Garrett Lisa Clemans-Cope Linda Blumberg Aaron Lucas Paul Masi Pool Administrators. Inc. Karl Ideman, President Richard Larose, Vice President Actuarial Research Corporation Gordon R. Trapnell, President Jim Mays, Vice President Consultants Katherine Swartz, Hvd. Sch. Pub. Hlth Patrick Collins, American Re.

  3. Roadmap • Overview of progress since Sept meeting • Roles: modeling and state consultation • Modeling: data base construction & policy simulation • Modeling: generic & state-specific versions • Face-validity survey, other state interactions • Some lessons learned (Emerging state results in later session)

  4. Progress Since September • Generic model construction, Oct-June • RI, WA, WI selected to participate, Nov • Cyberseminar on “risk premium,” Dec • Environmental scan on prevailing benefit levels, regulations, market structure, Dec-Jan • Site visit and seminars in Madison, WI, Jan • “Field testing” premiums in three states, Mar-Apr • State-specific modeling, Jun-

  5. Focus of Modeling • Reinsurance Institute from start emphasized HealthyNY-style retrospective reinsurance • i.e., at end of year, cover high-cost claims that occurred within a specified corridor (X% of costs $Y - $Z) . . . . . . not prospective “ceding” of individuals’ entire cost (share of all claims for Jones) • Participating states reinforced this focus . . . but are very interested in helping the already insured, unlike HealthyNY

  6. Simulation Modeling Approach for Reinsurance • Use a generic modeling approach, then customize to state populations and circumstances • Goal 1: Estimate effects of reinsurance policy alternatives on: • Health insurance premiums • Employer offer of health insurance • Health insurance coverage • Health care expenditures • State program costs • Goal 2: Describe how effects are distributed: • By health status • By income • By firm size

  7. Reinsurance Policy Parameters Subject to Modeling • Parameters for retrospective reinsurance • Lower reinsurance threshold (i.e., attachment point) • Upper limit • Coinsurance retained by original insurer • Reinsurance eligibility parameters • Group market, nongroup market, or both • Firm size criteria (under 10, 10 to 24, 25-49, 50-99, 100+) • Income or wage brackets • Rating rules • Nongroup pooling follows state rules • Can vary by age, health status, gender, single/family, family composition of age/gender/health status

  8. Why is Reinsurance challenging to model at the state level? • Data need to be population-based, representative of state health expenditures and insurance coverage, demographic, employer characteristics • Need an accurate distribution of health expenses in the upper tail of the distribution across risk groups  sample size concerns Problem: There are no state datasets with those features. Solution: Build a new state dataset suited to the task!

  9. The two main parts of the simulation model First, we model the baseline current state of insurance in State X. Then, we model the effect of reinsurance on the baseline.

  10. Baseline Overview: From National to State-specific Microdata • Medical Expenditure Panel Survey-Household Component (MEPS-HC) 1. Start with national data • Make expenditures consistent with National Health Accounts and high-cost claims data 2. Adjust health expenditures • Re-weight national data to match State X’s population characteristics 3. Benchmark To State X • Assign workers to synthetic employers to match State X’s firm size/industry mix in Statistics of U.S. Businesses 4. Create synthetic establishments • Build up premiums from covered expenses and benchmark to state data 5. Impute premiums Baseline database for State X is ready for simulating reinsurance programs

  11. Reweighted rates of insurance coverage for Rhode Island closely match state’s CPS data

  12. Reweighted distribution of family income for Rhode Island closely matches state’s CPS data

  13. Total annual private health expenditures and % of population by expenditure group Total private health expenditures (in mill$) Percent of population $30k-$50k $10k-$15k $15k-$30k Up to $2.5k $5k-$10k $2.5k-$5k $50k and up Expenditure groups Source: Reweighted MEPS-HC merged data 2001-2003. Final reweighted health expenditures for Wisconsin

  14. Construction of initial premiums • Employer Sponsored Insurance (ESI) premiums • Blend covered expenses from an employer group with an overall book of business = “pure premium” • Apply administrative loading factors by firm size • Compute establishment-level premium as the weighted average of single and family premiums within establishment’s risk group • Nongroup premiums • Establish rating cells for people with nongroup coverage • Blend covered expenses and predicted covered expenses • Apply administrative loading factor for nongroup coverage

  15. Baseline data including initial premiums 1. Specify reinsurance policy parameters 5. Compute ESI/NG premiums for new risk pool 2. Recompute ESI/NG premiums 3. Compute ESI offer changes 4.Compute changes in “take-up” of ESI/NG Application of reinsurance and flow of simulation model

  16. Premium Elasticity Targets • Elasticity = %change in rate / %change in premium • Employer elasticity of health insurance offer • Establishments with 25-99 workers: -0.4 • Stronger effects for smaller employers • Employee take-up elasticity in offering firms • Single coverage: -0.015 • Family coverage: -0.03 • Non-group coverage elasticities • Single coverage: -0.6 • Family coverage: -0.3

  17. Other Behavioral Assumptions • Medicaid and other public coverage unaffected • Employees and dependents will not drop ESI for non-group coverage • Dependents with ESI from a policy holder over age 64 or a government employee retain that coverage • ESI policy holders in families with more than one policy holder retain their original coverage • Non-group coverage is available to all at prevailing premiums within rating cell • Full value of subsidy passed back in the form of lower premiums

  18. Key Modeling Limitations • Uncertainty remains in the magnitude of behavioral effects • Model does not capture • Administrative costs, certain policy details, implementation issues • Benefit design issues, e.g. provider network • Certain insurer responses, e.g. number of insurers entering/exiting market • Insurer’s underwriting behavior • Leaves a significant role for qualitative analysis

  19. Questions the Model Can Address • What are the overall effects on premiums, coverage, costs? • Who goes where? • Where are the benefits (i.e., reduced premiums) concentrated? • How do risk pools change? • How does the composition of the uninsured change?

  20. “Field Test” Validation • Goal: establish “face validity” of modeling • Information from 3 states, 4 market segments • nongroup (a.k.a. individual) • small group (by state def’n) -- big focus of interest • large group (50+) • state employees • Variety of sources • Existing insurer surveys, by state, by high-risk pools • Web postings • Interviews • Found reasonable consistency of state-level data with MEPS-IC premiums and adjusted model estimates

  21. Some Lessons Learned • Policy starts best with clarification of goals and just how option(s) meet them • Data bases and programming take time • Policy makers and stakeholders think in terms of existing programs & products… • …whereas model relies more on demographic and other characteristics of people & firms • Achieving consistency with state indicators (e.g., in number of uninsured) helps buy-in and strengthens policy relevance

  22. End

More Related